For a couple of years, I’ve been toiling on a pet project : a personal anime database. The original idea was to re-use the extensive voice-acting data from ANN, and re-organize it so that I could browse it in creative ways. In particular, for each new anime I’d watch, I could get a quick overview of where I’d heard those voice-actors before ; filtered by the series I had actually seen, and in chronological order, unlike the messy ANN presentation.
Then this year, I did a number of major updates that pushed the project into a completely different dimension :
- I finally got around to getting the hang of HighCharts, which allowed for eye-catching graphs that made the data pop in very interesting ways.
- I overhauled the tagging system, allowing for better search and a bunch of new tag categories for more insight on what I’ve been watching (studios, adaptation source, is the show taking place in Japan and involving high-schoolers…). Of course, this was months before MAL started redesigning itself so that a lot of such data would be much easier to find.
- And most importantly, I added the series I hadn’t watched to the 500+ I had, quadrupling the database’s size in the process. This included a back-tagging ordeal that took me months and is only barely close to being over (with 100ish shows from the 70s left).
The net result, though, is that I now have a very searchable and malleable source of data to learn more about the history of anime, its fads and its evolution. Sure, there are some blind spots (I excluded OVAs, movies, shorts, and a good number of kids’ shows), but you can see the 00s ballooning up until the 2008 crisis, and the industry recovering since, as well as the rise of the light-novel adaptation (which is slightly overstated ; they only represent less than a fourth of all full-length new anime series started over the last few years, as opposed to the 40+% still firmly held by manga adaptations).
Another welcomed side-effect of adding 1500+ anime series to this database has been that I had to look up all of them so that I could tag them properly… and thus I discovered many series I’d never heard of before, but now really want to see. The moment I realized that (and modified the engine so I could keep a handy list of them) was one of the most heart-warming of the year.
Sure, it isn’t a fool-proof process ; some of the shows thus selected have been stinkers (ugh, Zombie-Loan) or barely above average (Kanamemo, Manabi Straight…). But if the net result is that I can experience such out-of-left-field surprises as Moonlight Mile and a show I’ll be discussing in a couple of days, it’s worth it.
